False positive nested-min-max

[XuehaiPan]

Bug description

The nested-min-max suggests max(3, max(1, 2)) to be max(3, 1, 2). However, if the min / max function takes an iterable max(iterable) instead of multiple inputs max(a, b, c). The nested min/max cannot be rewritten as a single statement max(x, max(iterable)) -> max(x, iterable).

"""test.py"""

lst = [1, 2]
max(3, max(lst))

If rewrite it to max(3, lst), it will raise:

TypeError: '>' not supported between instances of 'list' and 'int'

Configuration

No response

Command used

pylint test.py

Pylint output

************* Module test
test.py:4:0: W3301: Do not use nested call of 'max'; it's possible to do 'max(3, lst)' instead (nested-min-max)

------------------------------------------------------------------
Your code has been rated at 5.00/10 (previous run: 5.00/10, +0.00)

Expected behavior

No warnings.

Pylint version

pylint 2.16.0
astroid 2.14.1
Python 3.8.13 (default, Oct 21 2022, 23:50:54)

OS / Environment

No response

Additional dependencies

No response

[clavedeluna]

I think this makes sense and follows the logic for pylint also not raising the message for these types of inputs:

# This is too complicated (for now) as there is no clear better way to write it
max(max(i for i in range(10)), 0)
max(max(max(i for i in range(10)), 0), 1)

[Pierre-Sassoulas]

What is wrong with max(lst + [3])?

[nickdrozd]

What is wrong with max(lst + [3])?

It allocates an extra list object, which might be undesirable.

How about splicing in the list?

lst = [1, 2]
max(3, *lst)

[solney-syn]

Splatting the list (or allocating a new one) might be ways to avoid the nested min/max - but that's not what the fixer recommends. It says

it's possible to do 'max(3, lst)'

Which is incorrect in this case, no?

[nickdrozd]

Maybe it could suggest max(3, *lst) if the value of lst could be inferred to be an iterable? Is that feasible?

[vejkse]

I’m not sure I should open a new issue or not, but the current solution (in pylint 2.17.0) still gives wrong suggestions in some cases.

With

max(3, max([5] + [4]))

we get the correct suggestion

test.py:4:6: W3301: Do not use nested call of 'max'; it's possible to do 'max(3, *[5] + [4])' instead (nested-min-max)

but with

max(3, max([5] + [i for i in range(6,10) if i % 2 == 0]))

we get the incorrect one

test.py:5:6: W3301: Do not use nested call of 'max'; it's possible to do 'max(3, [5] + [i for i in range(6, 10) if i % 2 == 0])' instead (nested-min-max)

[Pierre-Sassoulas]

@vejkse thank you for reporting. Can you open a new issue please ? It's going to get lost in a closed issue otherwise.